Urban Heat Island and Its Effect on Dweller of Kolkata Metropolitan Area Using Geospatial Techniques

International Journal of Computer Sciences and Engineering Open Access Research Paper Vol.-6, Issue-10, Oct 2018 E-ISSN: 2347-2693

Urban Heat Island and its effect on Dweller of Kolkata Metropolitan area using Geospatial Techniques

Md Amir Ali Gazi 1, Ismail Mondal2

1Dept. of Geospatial Science, Burdwan University, India 2Dept. of Marine Science, University of Calcutta, 35 Bc Road, Kol-19, India

*Corresponding Author, e-mail: [email protected], Mobile No. +919836952240

Available online at www.ijcseonline.org

Accepted: 20/Sept/2018, Published: 31/Oct/2018 Abstract- The significance of Urban heat island (UHI) is to space heater than the urban adjacent area. This heat energy creates by urban people, house, shops, cars, buses, trains, industrial zone, etc. The UHI be an vital occurrence of urban environment in addition to gives direct and circumlocutory effect lying on urban population. In UHI calculating using landsat thermal band, Landsat TM, ETM+ and OLI satellite image (2000, 2008, 2017 Year) data processed to obtain a atmoshphire window method use to retrieve the land surface temperature (LST) using specifically band-6 (TM, ETM+) and band 10 (OLI) for data procurement and investigation. UHI consequence related with rising impermeable surface both spatially and temporally. This study aims to analyze the changes in LST with advent of the Kolkata Metropolitan Area (KMA). The result found that the (LST) is increasing sharply. The mean temperature of the area was 22.33°C in 2000 which became 23.68°C in 2008 and 23.79°C in 2017. The relation of built-up (NDBI) and LST is found positively correlated with a r value of 0.96 in 2000 and 0.78in 2017 and the relation with vegetation (NDVI) is negatively related and the r value is -0.98 in the years of 2000 and the r value is -0.97 in 2017, several heat zones are highlight and have been identified on the map of the KMA area. The addition of Geospatial technology be set up in the direction of valuable in monitor and analyze urban expansion pattern and in evaluation urbanization collision on surface temperature. Finally, the study suggests considering the possible micro-climatic changes in Kolkata metropolitan area and planning for the sustainable improvement.

Keyword: Urban Heat Island (UHI), land surface temperature (LST), Normalized differential build up index (NDBI), Normalized differential vegetation index (NDVI).

I. INTRODUCTION The present study of spatial spreading out of urban built-up areas of Kolkata Metropolitan areas always needs precise The Kolkata metropolitan area is a planning area, formed information on the size, shape and spatial construction of 1970 by metropolis development organization by the built-up features [22-24]. Therefore, a scientific and administration of west Bengal. It mechanism at the present vigorous method is necessary to rapidly retrieve such in condition of the west Bengal municipality and country information, temporal resolutions, consistent and repetitive (development and improvement) act, 1979 its planning measurements and synoptic views of Land Use and Land directorate was set up 1974[1-2]. The Land Surface Cover (LULC) [22, 25-28].These techniques can be broadly Temperature (LST) images consequent from space borne grouped into three categories: (a) pixel-based classification; satellite sensors have been used for Urban Heat Island (b) object-based classification [29-30] and (c) application (UHI) studies. The present study utilizes thermal data for of spectral indices such as Normalized Difference Built-up mapping of UHI [3-8]. Landsat TM, ETM+ and OLI data Index (NDBI)[31-35]. For the calculating NDVI and have also been used to analyses the effect of different Normalized Difference Built-up Index (NDBI) were degrees of urbanization on LST and UHI [9-17]. Other modified by changeover of TM bands with corresponding thermal sensors such as Advanced Spaceborne Thermal ETM+ bands. Finally, the satellite image was obtained by Emission and Reflection Radiometer (ASTER), [18-19] simple arithmetic process on enhanced NDBI, modified National Oceanic and Atmospheric Administration- NDVI and modified NDBI images of incessant type [34, Advanced Very High Resolution Radiometer (NOAA- 36-37]. AVHRR) and Envisat Advanced Along-Track Scanning Radiometer were also used for urban LST and interrelated The UHI of Kolkata city region which is significance that application [20-21]. heat heater than the urban nearby area [38, 15]. The perception of UHI consider the air temperature distinction

© 2018, IJCSE All Rights Reserved 741 International Journal of Computer Sciences and Engineering Vol.6(10), Oct 2018, E-ISSN: 2347-2693 among a city center and the surrounding area, UHI depends image and pixel size 30 × 30 meters (Table1). We are using on environmental phenomena that are Land Surface the LST data from Landsat TM, ETM+ Band 6 (TIR) and Temperature (LST). The LST is depends on isolation as OLI band 10 (TIR) are geometrically corrected, and well as nature of the surface or object materials that is conducted from Arc GIS and Erdas Imagine software and vegetation area, water bodies, soil, sand, and built up area. LST data thermal infrared (TIR) were radio metrically and So, the present study area we are found that there is positive geometrically corrected images to elaborate satellite relation between LST and urbanization. Accordingly, temperature calculate and emissivity for the image year of acquire LST is the most important and explanation pace to 2000, 2008 and 2017. At first, the DN value of Landsat TM the urban heat island analysis [39]. Finally, the study area band was transformed in to spectral radiance [41-42]. reveals to estimate that Land surface temperature (LST) for Lλ = ((LMAXλ - LMINλ)/(QCALMAX-QCALMIN)) * the year of 2000, 2008 and 2017 to correlate Normalized (QCAL-QCALMIN) + LMINλ[43] Differential Vegetation index (NDVI) and Normalized Where: Differential Built up Index with Land Surface Temperature Lλ = Spectral Radiance at the sensor's aperture in watts/ to suggest considering the potential of micro-climatic (meter squared * ster * µm) changes in KMA and planning for the sustainable LMAXλ = the spectral radiance that is scaled to development. Every day planners use geospatial technology QCALMAX in watts/(sq. m) to research, build up, implement, and observe the procedure QCALMIN = the minimum quantized calibrated pixel of their plans. Urban planners and decision-makers have to value (corresponding to LMINλ) in DN be more susceptible in the decision-making process and by QCALMAX = the maximum quantized calibrated pixel analyzing qualified data, to make the best decisions. Finally value (corresponding to LMAXλ) in DN. we attempted that to specific objectives are to assess Step - II metropolitan expansion patterns in the KMA and to analyze Next spectral radiance has been converted to Temperature the collision of the urban expansion on surface temperature. in Kelvin by formula is

II. MATERIALS AND METHOD

A. Study Area Where: The present study of urbanized area abstraction was led on T = Effective at-satellite temperature in Kelvin KMA; geographical location is 22.36N to 23.08N latitude to K2 = Calibration constant 2 from metadata 88.39E to 88.42E longitude (Figure. 1) and total 1010 sq km K1 = Calibration constant 1 from metadata area and capital of West Bengal. To settled mainly along the Lλ = Spectral radiance in watts/ (sq. m) banks of the river Hooghly about 150 km to the north of the Next Kelvin converted in Temperature (C = K – 273.15) Bay of Bengal right over the Gangetic delta plains [24]. and get the Temperatures. This Land surface temperature This is one of the important built-up areas in India, as well are closely related vegetation and built up area so their as of the urban world characterized by swift urbanization calculation this model NDVI and NDBI. and related spatio-temporal difference in biophysical Where, practices. KMA, the largest metropolitan accumulation in The NDVI is calculated from the equation eastern India, extending over 1851.41 km2 comprises 4 NDVI = (NIR – Red) / (NIR + Red) municipal corporations, 39 municipalities, and 1 Where cantonment, and parts of 24 panchayat samiti (rural local NDVI = (NIR – RED) / (NIR +RED) governments at the in-between equal in governmental In this Landsat 5 and 7 Image = (band 4 – band 3) / (band 4 construction of India) [24]. Kolkata is the most essential + band 3) commercial and industrialized center of east and north-east NDVI = Normalized Differential vegetation index Indian vicinity, and it holds essential manufacturing and NIR = Near Infra-Red band and RED = Red band transportation infrastructure [40]. The population of core NDBI = (SWIR – NIR) / (SWIR + NIR). Kolkata was 1.5 million in 1901, 11 million in 1991 and an In this Landsat 5 and 7 Image = (band 5 – band 4) / (band 5 exceptional 14.2 million in 2011[24]. + band 4). NDBI = Normalized Differential built up index B. Land surface temperature (LST) Formation SWIR = Short wave Infra-Red, NIR = Near Infra-Red The present study area collected Landsat 5 TM, Landsat 7 Calculation different year result Correlation between LST ETM+ and 8 OLI satellite image year of 2000, 2009 and and NDVI and NDBI [43].

2017 in month of February, downloaded from “Earth III. RESULT AND DISCUSSION Explorer” website of United States Geological Survey (USGS), path 138, and row 44. This image universal A. Spatial Temperature Change transverse projection (UTM) (with zone 45 north) and In this LST map represent that Spatial Temperature on the World geodetic system (WGS) – 1984 datum applied this surface; the author founded that in year of 2000 maximum

© 2018, IJCSE All Rights Reserved 742 International Journal of Computer Sciences and Engineering Vol.6(10), Oct 2018, E-ISSN: 2347-2693 temperature 30°C at central Kolkata Municipal Corporation The results show that the strong, negative correlation area and Kalyani and that is the build and industrial area, between LST, NDVI and NDBI implies that the higher and minimum temperature 10°C at Maheshtala and Budge biomass a land cover has, the lower the surface temperature. Budge municipal area that is the fellow land. Temperature Because of this relationship between LST, NDVI and NDBI variation is 20°C and means LST was 22.33°C and standard have an indirect impact on surface temperatures through deviation 1.21°C (Figure. 2 and Table 2). In year of 2008 NDVI. The values of NDVI in every year were scaled found that maximum temperature 33.12°C founded at according to the subsequent matrix, as the complete values central Kolkata Municipal Corporation area and Howrah of NDVI tend to very temporally in a non-systematic Municipal Corporation and Kalyani and Dum Dum manner [44] . Municipal area that is the build and industrial area, and minimum temperature 20.26°C at Uluberia, Budge Budge, C. Different yearly Land Surface Temperature Change Sonarpur, Chanddanagar Municipal area and average The impact of land features of surface temperature preserve temperature 23.68°C and standard deviation 1.28°C (Figure. as well to observe spatially. The surface temperature change 3 and Table 2). Year of 2017 found that maximum image obtained by image different is recorded into KMA temperature 35.00°C founded at central Kolkata Municipal area zones on the digital techniques method of equal Corporation area and Howrah Municipal Corporation and interval. The interrelation correlation matrix change results Kalyani and Dum Dum Municipal, Barrackpur and Garuliya to analysis, the different years; the LST maps of 2000 to municipal area that is the build and industrial area, and 2008 was -12.09°C and 5.09°C (Figure. 7) and year of 2008 minimum temperature 14.37°C at Uluberia, Budge Budge, to 2017 was -7.91°C and 9.64°C (Figure. 8) and year of Sonarpur, Baidyabati, Chapdani and Barasat Municipal 2000 to 2017 was -12.90°C and 7.15°C (Figure. 9). The corporation area that are the agricultural land and fellow Land Surface Temperature Change of Kolkata Metropolitan land average temperature 23.35°C and standard deviation area from 2008 to 2017 it is increase more than 2000 to 1.47°C (Figure. 4 and Table 2). 2008 and 2000 to 2017 map cumulative effect. At this time build up area road, buildings and industrial area increase B. Correlation Matrix between LST, NDVI and NDBI maximum. Such as this change LST is creating Urban Heat The correlation among LST, NDVI and NDBI was explored Island like other area. The change of LST was Maximum for every built up area during relationship analysis (pixel to 7.15°C to 9.64°C in area of Kolkata municipal Dum Dum, pixel). Figure 6 and Table 3-5 shows the correlation Barrackpur, Kalyani and Kancharapara Municipal area are coefficients among the two fundamentals in 2000, 2008 and respectively. The GIS method and correlation analysis 2017. The significance of each correlation coefficient was between the temperature zones of urban expansion of hit determined using a regression matrix. It is apparent from island to gives multiple r value 0.98 in the years of 2000 table 3-5 that LST values be inclined to depressingly and the r value is -0.97 in 2017, thus the most important to associate NDVI values for all land surface types in both the conclusion that spreading out is contributing to the years. amplify in surface temperature.

The correlation matrix results the year of 2000 where NDVI IV. CONCLUSIONS maximum 0.56 their temperature 9.95°C and whereas NDVI value -0.49 there LST 30°C and (r = -0.98). In the year of Beyond the study and analytical work, it is able to be 2008 whereas NDVI maximum 0.57 their temperature accomplished that UHI phenomenon is connected with the 20.26°C and where NDVI value -0.37 there LST 33°C and urbanization progression. The method of assessable remote (r = -0.97). The year of 2017 whereas NDVI maximum 0.67 sensing and GIS vis-s-vis digital image processing matrix, their temperature 14.37°C and where NDVI value -0.99 correlation and regression analysis, remote sensing and GIS there LST 35°C (r = -0.97). The relation between LST and technique validate that urbanized area increase to built-up NDVI negative and correlation between NDVI and LST stretch has through influence over the ecological have shown (Figure. 5 and Table 3-5). possessions. The correlation matrix results the year of 2000 where NDBI maximum 0.04 their temperature 30.00°C and NDVI value - The amalgamation of geospatial technology provides a 0.95 the value of LST 9.95°C and (r = 0.96). The year of resourceful method to sense urban growth of hit island and 2008 where NDBI maximum 0.50 their temperature also appraise its collision on surface temperature. The DIP 33.00°C and NDBI value -0.61 there LST 20.26°C and (r = techniques attached with GIS has established its aptitude to 0.94). The year of 2017 whereas NDVI maximum 1their present inclusive information on the environment, velocity temperature 35.00°C and where NDVI value -1 there LST and position of urban land expansion. The UHI effect has 14.37°C and (r = 7.8). Finally, the relation between LST been more prominent core area of Dum Dum, Barrackpur, and NDBI positive and LST change increase in the Kalyani and Kancharapara Municipal area are respectively. maximum build up area have been shown (Figure. 6 and It's in addition to experiential that the urban expansion Table 3-5). introduces UHI concentration during the progression of

© 2018, IJCSE All Rights Reserved 743 International Journal of Computer Sciences and Engineering Vol.6(10), Oct 2018, E-ISSN: 2347-2693 degradation of vegetation covers. The study reveals that the [15]. Mohan M, Kikegawa Y, Gurjar B, Bhati S, Kandya A, Ogawa K. remote sensing image is ideal for analyzing the Urban Heat 2012. Urban heat island assessment for a tropical urban airshed in Island effect of last 17 years ‘span. The satellite data India. Atmos Clim Sci. 2:127–138 [16]. Pandey P, Kumar D, Prakash A, Masih J, Singh M, Kumar S, analyzing and interpretation results shows the different Jain VK, Kumar K. 2012. A study of urban heat island and its years we found that development of Calcutta Metropolitan association with particulate matter during winter months over Area growth building and road and industries causes for Delhi. Sci Total Environ. 414:494–507. Land Surface Temperature Increase in Kolkata and [17]. Pichierri M, Bonafoni S, Biondi R. 2012. Satellite air temperature surrounding area a micro climate heat zone. So needed a estimation for monitoring the canopy layer heat island of Milan. Remote Sens Environ. 127:130–138. prior attention is needed to address the micro-climate [18]. Nichol J. 2009. An emissivity modulation method for spatial changes of the area for a sustainable city development. enhancement of thermal satellite images in urban heat island Finally, to observe the ecological crash of urban growth, the analysis. Photogramm Eng Remote Sens. 75:547–556. map pattern of ecological change can be associated to urban [19]. Mallick J, Rahman A, Singh CK. 2013. Modeling urban heat growth pattern by correlation analysis. islands in heterogeneous land surface and its correlation with impervious surface area by using night-time ASTER satellite data in highly urbanizing city, Delhi-India. Adv Space Res. 52:639– REFERENCES 655 [20]. Stathopoulou M, Cartalis C. 2009. Downscaling AVHRR land [1]. T. V. Ramachandra & Bharath H. Aithal & M. V. Sowmyashree surface temperatures for improved surface urban heat island (2014). Urban structure in Kolkata: metrics and modelling intensity estimation. Remote Sens Environ. 113:2592–2605. through geo-informatics, Appl Geomat DOI 10.1007/s12518- [21]. Badarinath K, Chand TK, Madhavilatha K, Raghavaswamy V. 014-0135-y, # Società Italiana di Fotogrammetria e Topografia 2005. Studies on urban heat islands using envisat AATSR data. J (SIFET), pp.1-15. Indian Soc Remote Sens. 33:495–501. [2]. Chakraborty, K. R. (1991). “Urbanisation Trend in the Calcutta [22]. Xu H. 2008. A new index for delineating built‐up land features in Metroplitan District”, in Dasgupta, B. et al. (eds.) “Calcutta's satellite imagery. Int J Remote Sens. 29(14):4269–4276. Urban Future- Agonies From The Past and Prospects”, DOI:10.1080/01431160802039957. Government of West Bengal, Calcutta, pp. 114. [23]. Wang Z, Gang C, Li X, Chen Y, Li J. 2015. Application of a [3]. Mallick J, Kant Y, Bharath B. 2008. Estimation of land surface normalized difference impervious index (NDII) to extract urban temperature over Delhi using Landsat-7 ETM+. J Indian Geophys impervious surface features based on Landsat TM images. Int J Union. 12:131–140. Remote Sens. 36(4):1055–1069. [4]. Ambinakudige S. 2011. Remote sensing of land cover’s effect on [24]. Khan A, Chatterjee S, Akbari H, Bhatti SS, Dinda A, Mitra C, surface temperatures: a case study of the urban heat island in Hong H & Doan QV (2017): Step-wise Landclass Elimination Bangalore, India. Appl GIS. 7:1–12. Approach for extracting mixed-type built-up areas of Kolkata [5]. Bajaja, DN, Inamdarb, AB, Vaibhava, V. 2012. Temporal megacity, Geocarto International, DOI: variation of urban heat island using Landsat data: a case study of 10.1080/10106049.2017.1408704. Ahmadabad, India, in: Aiming Smart Space Sensing. Presented at [25]. Guindon B, Zhang Y, Dillabaugh C. 2004. Landsat urban the 33rd Asian Conference on Remote Sensing; Thailand mapping based on a combined spectral–spatial methodology. [6]. Borthakur M, Nath BK. 2012. A study of changing urban Remote Sens Environ. 92(2):218–232. DOI: landscape and heat island phenomenon in guwahati metropolitan 10.1016/j.rse.2004.06.015. area. Int J Sci Res. 2:1–6 [26]. Bhatta B. 2009. Analysis of urban growth pattern using remote [7]. Kumar KS, Bhaskar PU, Padmakumari K. 2012. Estimation of sensing and GIS: a case study of Kolkata, India. Int J Remote land surface temperature to study urban heat island effect using Sens. 30(18):4733–4746. Landsat ETM+ image. Int J Eng Sci. 4:771–778 [27]. Griffiths P, Hostert P, Gruebner O, Sebastian VDL. 2010. [8]. Sharma R, Ghosh A, Joshi PK. 2013. Spatio-temporal footprints Mapping megacity growth with multi-sensor data. Remote Sens of urbanisation in Surat, the diamond city of India (1990–2009). Environ. 114(2):426–439. DOI: 10.1016/j.rse.2009.09.012. Environ Monit Assess. 185:3313–3325. [28]. Bouzekri S, Lasbet AA, Lachehab A. 2015. A new spectral index [9]. Hung T, Uchihama D, Ochi S, Yasuoka Y. 2006. Assessment for extraction of built-up area using Landsat-8 data. J Indian Soc with satellite data of the urban heat island effects in Asian mega Remote Sens. 43(4):867–873. cities. Int J Appl Earth Obs Geoinf. 8:34–48. [29]. Qian J, Zhou Q, Hou Q. 2007. Comparison of pixel-based and [10]. Pongracz R, Bartholy J, Dezso Z. 2006. Remotely sensed thermal object-oriented classification methods for extracting built-up information applied to urban climate analysis. Adv Space Res. areas in Aridzone. In: ISPRS Workshop on Updating Geo-spatial 37:2191–2196. Databases with Imagery & The 5th ISPRS Workshop on [11]. Tran H, Uchihama D, Ochi S, Yasuoka Y. 2006. Assessment DMGISs; pp. 163–171). with satellite data of the urban heat island effects in Asian mega [30]. Cleve C, Kelly M, Kearns FR, Moritz M. 2008. Classification of cities. Int J Appl Earth Obs Geoinf. 8:34–48. the wildland–urban interface: a comparison of pixel- and object- [12]. Imhoff ML, Zhang P, Wolfe RE, Bounoua L. 2010. Remote based classifications using high-resolution aerial photography. sensing of the urban heat island effect across biomes in the Comput Environ Urban Syst. 32(4):317–326. DOI: continental USA. Remote Sens Environ. 114:504–513. 10.1016/j.compenvurbsys.2007.10.001. [13]. Peng S, Piao S, Ciais P, Friedlingstein P, Ottle C, Bréon FM, [31]. Zha Y, Gao J, Ni S. 2003. Use of normalized difference built-up Myneni RB. 2011. Surface urban heat island across 419 global index in automatically mapping urban areas from TM imagery. big cities. Environ Sci Technol. 46:696–703. Int J Remote Sens. 24(3):583–594. [14]. Guo Z, Wang SD, Cheng MM, Shu Y. 2012. Assess the effect of DOI:10.1080/01431160304987. different degrees of urbanization on land surface temperature [32]. Zhang Q, Pavlic G, Chen W, Fraser R, Leblanc S, Cihlar J. 2005. using remote sensing images. Procedia Environ Sci. 13:935–942. A semi-automatic segmentation procedure for feature extraction

in remotely sensed imagery. Comput Geosci. 31(3):289–296. [44]. Wang Z, Gang C, Li X, Chen Y, Li J. 2015. Application of a DOI: 10.1016/j.cageo.2004.10.003. normalized difference impervious index (NDII) to extract urban [33]. He C, Shi P, Xie D, Zhao Y. 2010. Improving the normalized impervious surface features based on Landsat TM images. Int J difference built-up index to map urban built-up areas using a Remote Sens. 36(4):1055–1069. semiautomatic segmentation approach. Remote Sens Lett. 1(4):213–221. DOI:10.1080/01431161.2010.481681. Author Profile [34]. Bhatti SS, Tripathi NK. 2014. Built-up area extraction using Landsat 8 OLI imagery. GISci Remote Sens. 51(4):445–467. Mr. Md Amir Ali Gazi, Msc. Department of Geospatial DOI:10.1080/15481603.2014.939539. Science, Burdwan University, India. [35]. Lu N, Hernandez AJ, Ramsey RD. 2015. Land cover dynamics monitoring with Landsat data in Kunming, China: a cost- Dr. Ismail Mondal, Post-Doctoral Fellow effective sampling and modelling scheme using Google Earth Department of Marine Science, University of Calcutta, imagery and random forests. Geocarto Int. 30(2):186–201. Kolkata-700019, India, Mobile No. +919836952240 [36]. Ke Y, Im J, Lee J, Gong H, Ryu Y. 2015. Characteristics of Landsat 8 OLI-derived NDVI by comparison with multiple e-mail: [email protected], satellite sensors and in-situ observations. Remote Sens Environ. 164:298–313. [37]. Piyoosh AK, Ghosh SK. 2017. Development of a modified bare soil and urban index for Landsat 8 satellite data. Geocarto Int: 1– 20. [38]. Ahamed B, 2013. “Simulating Land Cover Changes and Their

Impacts on Land Surface Temperature in Dhaka, Bangladesh” www.mdpi.com/journal/remotesensing. [39]. Bhattacharjee D, Hazra S 2014. Distribution of Land Surface Temperature over Built-up Area by Web-GIS Techniques, Indian Journal of Spatial Science Vol - 5.0 No. 2 Winter Issue 2014 pp. 70 – 76. [40]. Nath SK, Adhikari MD, Devaraj N, Maiti SK. 2015. Seismic vulnerability and risk assessment of Kolkata City, India. Nat Hazards Earth Sys Sci. 15(6):1103–1121. [41]. Chander, G., Markham, B.L., Helder, D.L. (2009). Summary of Current Radiometric Calibration Coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI Sensors. (In Press, Remote Sensing of Environment, Manuscript Number: RSE-D-08- 00684)https://ntrs.nasa.gov/search.jsp?R=20090027884 2018-05- 09T11:26:42+00:00Z. [42]. Voogt JA, Oke TR, 2003. Thermal remote sensing of urban climates, Remote Sensing of Environment, Volume 86, Issue 3, 15 August 2003, Pages 370 384. [43]. Lin liu and Yuanzhi Zhang e, Urban Heat Island Analysis Using the Landsat TM Data and ASTER Data: A Case Study in Hong Kong, Remote Sensing, 2011, 3, pp1535-1552.

Table 1 details of Landsat data used (Sources: USGS Earth Explorer) Sl.No Satellite Sensor Date of acquisition 1 Landsat 5 TM 01/12/2000 2 Landsat 7 ETM+ 17/11/2008 3 Landsat 8 OLI 02/12/2017

Table 2 Changing LST in KDMA area Land Surface Temperature (LST) in °C Year Maximum Minimum Range Mean Standard Deviation 2000 30 10 20 22.33 1.21 2008 33.12 20.26 12.86 23.68 1.28 2017 35 14.37 20.63 23.35 1.47

Table 3 Correlation (r) among NDVI, NDBI, and LST, 2000 Correlation NDVI NDBI LST NDVI 1 NDBI -0.98 1 LST -0.98 0.96 1

Table 4 Correlation (r) among NDVI, NDBI, and LST, 2008

Correlation NDVI NDBI LST NDVI 1 NDBI -0.98 1

LST -0.97 0.94 1

Table 5 Correlation (r) among NDVI, NDBI, and LST, 2017

Correlation NDVI NDBI LST NDVI 1

NDBI -0.93 1

LST -0.97 0.76 1

Fig. 1. Location Map (Source: KDMA)

Fig.2. Land Surface Temperature Map 2000

Fig.3. Land Surface Temperature Map 2008

Fig.4. Land Surface Temperature Map 2017

Correlation between LST & NDVI 2017 40

C ° 20

LST 15 y = -28.992x + 31.214

10 R² = 0.9509

0 -0.4 -0.2 0 0.2 0.4 0.6 0.8

NDVI VALUE

Fig. 5.Correlation between LST and NDVI

Correlation between LST & NDBI 2017

y = 17.98x + 26.57

R = 0.784

45 40

C 30 ° 25

LST 20 15 10

0 -1.5 -1 -0.5 0 0.5 1 1.5 NDBI VALUE

Fig. 6. Correlation between LST and NDBI

Fig. 7. Land Surface Temperature Change Map of 2000 to 2008

Fig. 8. Land Surface Temperature Change Map of 2008 to 2017

Fig. 9. Land Surface Temperature Change Map of 2000to 2017